Two herb spice tobacco grinders were purchased from commercial retailers. The Cannabis research program at the National Institute of Standards and Technology provided 20 cannabis samples, all of which had the % total THC and % total CBD previously determined through Liquid Chromatography-Photodiode Array .First, 10 µL of the plant extract or reference standards were pipetted onto the PSPME substrate. Next, 10 uL of 0.1% FBBB solution was then pipetted onto the substrate followed by 10 uL of 0.1 N NaOH.The solvents evaporate within 1–2 min as the color develops. A red color is indicative of THC and an orange color is indicative of CBD. The FBBB test was performed in 5 replicates per extract. Each substrate was photographed with a Dino-Lite AM4115ZT digital microscope . A Dino-Lite AM4115T-GRFBY Digital Microscope was used to capture fluorescence images of the substrates. The Dino-Lite AM4115T-GRFBY uses a 480 nm excitation light source and contains emission filters for 510 nm and 610 nm. These images were taken in the absence of ambient light to remove interference from outside sources of light. The visible and fluorescence images are analyzed using the ImageJ software using the RGB measure plugin to obtain the average RGB numerical code across each substrate. All five blue ridge hemp samples and 20 cannabis samples of known cannabinoid concentrations were evaluated using the FBBB reagent with 5 replicates. For each replicate, a color image, a fluorescence image, and fluorescence spectra were obtained. The color results for all samples are summarized in Table 5. All of the hemp samples formed an orange color when reacted with FBBB, except for Sample 11 and Sample 12, which did not have any reaction. Of the 13 samples that are marijuana , 6 of them produced an orange color instead of the red color indicative of THC.
These six were Sample 6, Sample 9, Sample 10, Sample 18, Sample 19, and Sample 20. For samples 6, 9, 10, and 20 the total CBD was at a higher concentration than total THC, all containing a THC:CBD ratio below 1. Samples 18 and 19 had THC:CBD ratios of 1.0 and 1.4 respectively. The other marijuana type samples had a THC:CBD ratio much higher than 2 and formed a red color. Samples that had a THC:CBD ratio below 2 did not fluoresce brightly under the Dino-Lite microscope at 480 nm excitation. Importantly, the marijuana-type samples that either had no CBD or a high THC:CBD ratio did fluoresce brightly under the Dino-Lite at the same excitation. These results suggest that when there is more CBD than THC in the cannabis grow set up plant, or if the concentrations are similar, the FBBB will produce an orange color indicative of hemp rather than a red color indicative in marijuana. In addition, when the THC:CBD ratio is low, the fluorescence of the chromophore will also be low. The fluorescence spectra from the VSC2000 for hemp-type samples showed a low % intensity at 655 nm, typically between 10% and 20%, and a higher intensity at 695 nm, between 15% and 40%. The exception to this were samples 11 and 12 whose extracts did not react with FBBB and had similar spectra to the blank. The marijuana-type samples with a low THC:CBD showed similar spectra to the hemp samples, with fluorescence intensities at or below 20% at 655 nm for those with THC:CBD significantly lower than 1. Samples 18, 19, and 20, which have THC:CBD from 0.48 to 1.4, all showed slightly higher intensities at 655 nm than the hemp samples . For marijuana-type samples with a THC:CBD above 2, the intensity of fluorescence increases between 40 and 70% at 655 nm and 695 nm. Low fluorescence intensity for hemp samples at 655 nm is expected since there is very little THC in these samples. For samples 6, 9, and 10 there was much more CBD than THC in the cannabis plant leading FBBB + CBD to form over FBBB + THC. Samples 18, 19, and 20 showed a slightly more intense band at 655 nm. This increase could be attributed to the fact that there is a similar concentration of CBD and THC in these samples and allowed for FBBB to react with both THC and CBD.
In addition, all cannabis extracts contain a band at 695 nm. This interference is likely due to chlorophyll and other pigments from the plant material, however, even with this interference, the difference in fluorescence intensity between hemp and marijuana-type cannabis with a high THC:CBD is noticeable. When the THC:CBD ratio is below 2, the fluorescence intensity decreases. This is consistent with the results obtained from the color images and fluorescence images using the Dino-Lite microscopes. A comparison of a marijuana-type sample and a hemp-type sample through color images, fluorescence images, and the fluorescence spectra is shown in Fig. 7. Linear Discriminant Analysis was used as a supervised technique to determine whether FBBB can be used to correctly classify hemptype cannabis and marijuana-type cannabis. Each sample described in Table 5 was evaluated in 5 replicates. For each replicate RGB of the color image, RGB of the fluorescence image, and the % intensity at 655 nm and 695 nm in the fluorescence spectra were recorded. The LDA analysis was performed using the JMP software. The first LDA model was constructed using % intensity at 655 nm and % intensity at 695 nm values as the variables. The resulting model had an R2 of 0.61 and misclassified samples 6, 9, 10, 14, 15, 18, 19, 20. Samples 6, 9, 10, and 18–20 are marijuana-type samples with THC:CBD below 2, showing similar fluorescence spectra to hemp samples leading to their misclassification. Samples 6, 9, 10, and 18–20 were removed from the data set and LDA was performed again using the data from the 7 remaining marijuana-type samples and the 12 hemp-type samples. This analysis resulted in an R2 of 0.999 and no misclassifications. LDA was also performed using the R, G, and B codes for each color image and fluorescence image. LDA of all the samples using RGB for the color images produced an R2 of 0.51 and misclassified samples 3,4, 10, 18–20, and one replicate of 16 and 17 each. To improve the model, all marijuana type samples with THC:CBD below 2 were removed from the data set. Samples 11 and 12 were removed as well since they did not produce a color as they were likely the cause of the misclassification of samples 3 and 4, which produced a light red color.
This did improve the model with the R2 value of 0.95 and only misclassifying one replicate of sample 3. This indicates when using only RGB of the visible image, one should exclude samples that do not form a color as it may cause misclassification. An LDA model of all samples using RGB of the fluorescence images taken for each replicate was also made. This LDA model misclassified multiple hemp-type and marijuana-type samples resulting in an R2 of 0.46. When the marijuana type samples with THC:CBD below 2 were removed from the data set, there were no misclassifications and R2 was 0.995. Finally, an LDA model was made to classify the marijuana-type samples with a high THC:CBD and all the hemp-type samples using the R,G and B from the color images and R-F,G-F and B-F from the fluorescence images for a total of 6 variables. This model resulted in a clear separation between hemp-type and marijuanatype cannabis resulting in an R2 of 1.0 with G providing the highest correlation to hemp and R-F providing the highest correlation to THCrich cannabis . A Receiver Operating Characteristic of the model showed that the area under the curve for both hemp and marijuana are 1, displaying excellent selectivity and sensitivity when combining color and fluorescence to discriminate from hemp-type cannabis and marijuana type cannabis. The FBBB test was used to evaluate 6 different cannabinoids, 5 commercial hemp strains, 20 cannabis samples, and various herbs and spices. It was determined that when FBBB reacts with THC, it forms a red chromophore that fluoresces under 480 nm light. Conversely, when reacted with CBD or CBD-rich products, such as outdoor cannabis grow, an orange chromophore is formed, and this chromophore does not fluoresce. This is the first time, to the author’s knowledge, that the fluorescence of the FBBB + THC chromophore/fluorophore is reported for a colorimetric test. This fluorescence is easily visualized using a portable Dino-Lite microscope and its spectra obtained with a VSC2000 spectrometer. The intensity and wavelength of the fluorescence for the chromophore combined with the distinct red color it displays makes for a more selective and sensitive test to differentiate between marijuana and hemp. The structure for FBBB + THC has been previously determined by the Almirall lab, as shown in Fig. 1. The chromophore results from an extended conjugation of π-bonds decreasing the distance between energy transitions between the ground state and excited state.
This extended conjugation causes a “red shift” of the FBBB chromophore, which is responsible for the red color and the fluorescence that is observed when THC reacts with FBBB. One theory for CBD + FBBB lacking fluorescence intensity is that CBD has a less rigid structure than THC. It is known that structure rigidity and a fused ring structure increases the quantum efficiency, and therefore fluorescence of a molecule. Since CBD is less rigid than THC and does not have a fused ring structure, it is prone to relaxation through internal conversion rather than through radiative means. Therefore, FBBB + CBD likely relaxes through nonradiative mechanisms, which decreases overall fluorescence. The difference in both color and fluorescence that is observed for FBBB + THC and FBBB + CBD is an advantage that the FBBB test has compared to other tests for presumptive analysis of cannabis, which only use color. The selectivity of the FBBB test was evaluated by analyzing 5 other cannabinoids, herbs, spices, essential oils, tobacco, and hops. None of these substances produced color like that of FBBB + THC nor fluorescence observed. For the colorimetric calibration experiments, it was shown that when the ratio of THC:CBD is above 1, a red color forms indicating that there is marijuana present. These experiments also found that the absolute LODs for THC on the PSPME substrates was as low as 500 ng, which is significantly lower than the LOD for the D-L test . The THC LOD for the 4-AP test is not currently known but expected to be >500 ng. This study demonstrates that the FBBB test is very selective and sensitive for THC, forming a red color and an intense fluorescence that can be distinguished from other chromophores. In addition, this chromophore is long lasting, allowing the color and fluorescence to be observed long after the test is performed. This longlasting color is attributed to the nature of the FBBB being a diazonium salt, which are known to be stable and even used to form dyes in textiles. One limitation that was discovered for the FBBB test is that the reagent is not stable at room temperatures over more than a few days, losing its color and producing no reaction with THC or CBD. The FBBB reagent and the preloaded FBBB substrate were stable in the refrigerator/cooler for at least 45 days. The temperature instability is not ideal for field work since a kit using the Fast Blue BB test would likely be exposed to temperatures above 4 ◦C. For this reason, future work will focus on determining a method to maintain the FBBB stable at ambient temperatures. The analysis of the Blue Ridge Hemp and NIST samples demonstrate that FBBB is very effective at discriminating between hemp-type samples with THC content <0.3% and marijuana-type samples with a high THC content or THC:CBD ratios. Marijuana-type cannabis containing >0.3% THC and high CBD could be misclassified as hemp but these types of samples are uncommon in seized drugs. The results of these LDA models using RGB inputs support the observed findings of the visual evaluation of the Blue Ridge and NIST samples with FBBB.